SPSS Tutorials: Creating a Codebook

A codebook is a document containing information about each of the variables in your dataset, such as:

• The name assigned to the variable
• What the variable represents (i.e., its label)
• How the variable was measured (e.g. nominal, ordinal, scale)
• How the variable was actually recorded in the raw data (i.e. numeric, string; how many characters wide it is; how many decimal places it has)
• For scale variables: The variable's units of measurement
• For categorical variables: If coded numerically, the numeric codes and what they represent

Codebooks can also contain documentation about when and how the data was created. A good codebook allows you to communicate your research data to others clearly and succinctly, and ensures that the data is understood and interpreted properly.

Many codebooks are created manually; however, in SPSS, it's possible to generate a codebook from an existing SPSS data file.

To get the most out of the Codebooks procedure in SPSS, your dataset should already have variable labels and value labels applied before you run the Codebooks procedure. If you are not familiar with variable properties, such as labels or measurement levels, or concepts like value labeling of category codes in SPSS, you should read the Defining Variables tutorial before continuing.

Simple Codebook

This codebook method prints most of the information found in the Variable View window. It gives the names, labels, measurement levels, widths, formats, and any assigned missing values labels for every variable in the dataset. It also prints a table with the assigned value labels for categorical variables.

You can generate this simple codebook using the point-and-click menus, or using syntax.

Using the Menus

1. Open the SPSS datafile.
2. Click File > Display Data File Information > Working File.
3. The codebook will print to the Output Viewer window.

Using Syntax

DISPLAY DICTIONARY.

Detailed Codebook

This codebook method includes all of the same information as the simple method, but also includes options for printing summary statistics as well. Unlike the simple method, you can choose which variables are included in the codebook, and you can choose which variable properties are included in the summary. Also unlike the simple method, the summary information for each variable will be printed in its own table.

You can generate this detailed codebook using the Codebooks dialog window, or using syntax.

Note: This procedure was introduced in SPSS version 17 (source: SPSS v23 Command Syntax Reference). If you are using an older version of SPSS, this command is not available - it will not appear in the menus, and running the syntax will return error messages.

Using the Codebooks Dialog Window

1. Open the SPSS datafile.
2. Click Analyze > Reports > Codebook.
3. In the Variables tab: Add the variables you want in the codebook to the Codebook Variables box. To include all variables, click inside the Variables box, press Ctrl + A, then click the arrow button.
4. In the Output tab: (Optional) Choose what variable and datafile properties you want to be included in the codebook:
   1. Variable information: By default, includes Position, Label, Type, Format, Measurement level, Role, Value labels, Missing values, and Custom attributes.
   2. File information: None included by default.
   3. Variable display order: By default, ordered identically to how the variables are ordered in the file. Can also order alphabetically, by file, or by measurement level.
   4. Maximum number of categories: By default, limits to 200 categories.
5. In the Statistics tab: (Optional) Choose what statistics you want in the codebook. By default, counts and percents will be printed for nominal and ordinal variables, and mean, standard deviation, and quartiles will be printed for scale variables.
6. When finished, click OK.

Using Syntax

CODEBOOK <variables-names-here>
  /VARINFO POSITION LABEL TYPE FORMAT MEASURE ROLE VALUELABELS MISSING ATTRIBUTES
  /FILEINFO NAME CASECOUNT
  /OPTIONS VARORDER=VARLIST SORT=ASCENDING MAXCATS=200
  /STATISTICS COUNT PERCENT MEAN STDDEV QUARTILES.

Note: When listing the variable names in the syntax, the assigned measurement level must be given in brackets after each variable name: [s] for scale, [n] for nominal, [o] for ordinal.

To reproduce this example, download the sample SPSS dataset and SPSS syntax file. Run the syntax file on the sample data. This will add all of the appropriate variable labels and value labels for this dataset.

Problem Statement

When sharing your data with others, it's important that your variables are properly documented. This includes having succinct but descriptive labels for your variables, and labels for any numeric codes used for categories.

If you receive a dataset from a collaborator, you can get an overview of its contents by running the Display Dictionary procedure.

Running the Procedure

To generate a simple codebook for the sample data, click File > Display Data File Information > Working File.

Output

Syntax

DISPLAY DICTIONARY.

Tables

The first table is the Variable Information table. This table summarizes variable-level information, including:

• Variable name
• Position (i.e., the order of the columns)
• Variable label
• Measurement level (nominal, ordinal, scale)
• Role
• Column width
• Alignment
• Print format
• Write format

The second table is the Variable Values table. This table will only appear if you have value labels defined for at least one variable in your dataset; otherwise, it is omitted. This table prints the name of each variable with defined value labels, and lists each code and associated label for that variable.

Qualtrics users: This procedure works well with survey data that you've downloaded from Qualtrics in SPSS format. Use it to check the coding of your multiple choice items!